Generic Platform Technology Development: A breakthrough in Photonic Integration

Photonic Integrated Circuits (PICs) are considered as the way to make photonic systems or subsystems cheap and ubiquitous.  However, PICs still are several orders of magnitude more expensive than their microelectronic counterparts, and this has restricted their application to a few niche markets.  Paradigm targets a novel approach in photonic integration which will reduce the R&D costs of PICs by more than a factor of ten. It will bring the application of PICs that integrate complex and advanced photonic functionality on a single chip within reach for a large number of small and larger companies.  Europe presently has a world state-of-the-art position and is leading in this novel approach.

Moore’s law in Photonics

Figure 1 shows the complexity development of optical chips for application in Wavelength Division Multiplexing telecommunication systems, the technology that enabled worldwide high-speed internet and that is a major driver for photonic integration.  The complexity is measured as the number of optical components per chip.  Most points shown in the figure can be traced in earlier publications [1,2].  Figure 2 shows an early example of such a chip: an optical cross-connect (OXC) that can switch the four wavelength signals in both input ports independently between the two output ports.  The chip, integrates two wavelength demultiplexers with 16 optical switches on a chip area of 8x12mm2.  The most complex chip reported today is a WDM-transmitter chip which integrates 40 wavelength channels, each of them containing a laser, a modulator, a power monitor and a channel equalizer [3].

Figure 1 Component count in reported InP photonic integrated circuits over time.  The green (analogue) and red (digital) lines are projections discussed it the text.

Figure 1 shows a clear exponential trend, similar to Moore’s law in electronics, which suggests that Photonics is following the same process-driven development path as microelectronics, albeit at a slower pace and with a 30 years time shift.  Probably this trend is in response to the very same developments in process equipment which are driving Moore’s Law in Silicon technology.  There is an important difference, however.  The Micro-electronic ICs which support the relationship in Moore’s Law are commercially applied devices, whereas most of the complex Photonic ICs are research devices which ended in published work but did not make the transition to the marketplace.  The only truly complex chip which is currently applied in a commercial product is the WDM transmitter chip (10x10 Gb/s) of the US company Infinera [3], used in a 100Gb/s WDM system.  It is the first demonstration that complex PICs can provide a competitive edge in a commercial environment.

Figure 2: An optical cross-connect (OXC) chip that can switch 4 wavelength channels independently from two input waveguides to two output guides. It integrates 2 wavelength
demultiplexers with 16 optical switches on a chip area of 8x12mm2.

What went wrong?

It is an interesting question why so few of the advanced PICs reported in the literature have made it to the market, despite the fact that in the last two decades several billion dollars have been invested in development of integration technologies in national and international projects in Europe, America and the Far East.  The main reason is that they are too expensive to compete with other technologies like micro-optic or hybrid integration.  The problem with current project funding models within Europe is that they tie the technology development closely to an application: you get no money without a clear and challenging application.  In order to meet the challenging specifications the technology has to be fully optimized for that application and, as a result, we have almost as many technologies as applications.  Due to this huge fragmentation, the market for these specific technologies is usually too small to justify their further development into the industrial volume manufacturing process that would really lead to low chip costs.  This is quite different from the situation in micro-electronics where a huge market is served by a relatively small set of integration technologies (most of them CMOS technologies), and most of the technologies are used for a wide variety of applications.  In this way the development costs of the integration process are shared by a large number of applications and the volume of all applications together is sufficiently large to justify the development of a sophisticated industrial manufacturing facility for large volume production, which combines high performance with low fabrication costs.

The solution to the problem described above seems obvious: we should introduce to photonics the methodology that allowed microelectronics to change the world.  It is amazing that in photonics we are still developing expensive application specific technologies with poor market perspectives instead of developing low-cost Application Specific Photonic ICs in generic integration technologies that can serve a wide variety of applications and have much better market perspectives.

Some Definitions

The generic integration approach is expected to lead to a dramatic reduction of the costs of Photonic ICs, and a significant reduction of the number of design cycles needed to come to a satisfactory device.  This is mainly due to the fact that it offers access to a well characterized process, rather than simply a clean-room facility.  Presently, many companies are vertically integrated and bear the full cost of their fabs themselves.  However, a small number of companies is currently offering clean-room access to other so-called fabless customers.  In this business model such companies do process development for a customer’s devices.  This kind of foundry operation makes it possible to develop a product without having to build your own cleanroom, which leads to a significant capital cost reduction.  The process development costs, which can be in the range of several million Euros, are still specific for the customer’s product, however.  These companies are custom foundries.  In a generic foundry the costs of the process development are also shared by many users, and low fabrication costs can be realised at low volume.  Because the generic process is used by a large number of customers it is worth the effort of needed to create dedicated design kits with accurate models for the building blocks and powerful simulation engines.  Working in this way, PIC prototypes can be achieved cost effectively in a very small number of design and fabrication cycles.  This in itself will lead to a dramatic reduction of the cost of both PIC R&D and manufacturing.


[1] R. P. Nagarajan and M.K. Smit, June 2007, Photonic integration, IEEE LEOS Newsletter 21(3), pp4-10.

[2] M. K. Smit and C. Van Dam, PHASAR-based WDM devices: principles, design and applications, June 1996, J. Sel. Topics in Quantum Electron., 2(2), pp236-250

[3] R. P. Nagarajan et al., Feb. 2007 Large-scale photonic integrated circuits for long haul transmission and switching, J. Opt. Networking, 6(2), pp102-111.